The goal of this project is to elucidate the processes by which mothers transmit social knowledge and skills to their preschool children. During Year 01 of the project a laboratory procedure will be developed and refined. This procedure is designed to provide a context for the empirical examination of two distinct pathways of maternal social influence. These pathways arc conceptualized in terms of stylistic qualities of mother- child interaction (such as responsiveness) and the content of mothers' explicit coaching about peer interaction (such as the value of sharing). The mother-child assessment consists of (1) a naturalistic play session in which the mother is present in a small playroom with her child and a same- sex, unfamiliar peer, (2) a prompted coaching session in which the mother is asked to help her child interpret and determine an appropriate response to a series of videotaped peer-interaction problems, and (3) a teaching task in which the mother is asked to help her child solve a block-design puzzle. Ah original set of videotapes will be produced, each containing a series of vignettes depicting young children (child actors) in age- appropriate peer interaction situations. These vignettes will serve as stimuli for the assessment of maternal prompted coaching. The laboratory procedure will be tested in Year 01 on an economically and ethnically diverse sample of 40 mother-child pairs. Half of these mother-child pairs will complete the procedure twice in order to assess the stability of interaction and coaching patterns. Videorecords of the mother-child assessment will be coded for interactional style and coaching content. Measures derived from the coding then will be related to indices of children's classroom competence (peer and teacher ratings, observations, social problem solving). The laboratory procedure will be refined and tested further in Year 02 on a separate sample of 28 high-accepted, 28 low-accepted, and 28 average- status children and their mothers, drawn from a pool of 200 area preschool children. The same school measures will be collected as in Year 01. Comparability of measurement models and replicability of associations across studies will be examined. Analyses will explore the extent to which style and coaching variables operate additively, multiplicatively, or redundantly in the prediction of children's social behavior and social knowledge, and whether these associations are moderated by child sex and vocabulary IQ, and family SES. Findings from this project are expected to provide useful information on ways parents can participate in preventive interventions with preschool children at-risk for peer relationship problems.